Forecasting the next flu season

How bad could the next flu season be?
09 March 2021

Interview with 

Derek Smith, University of Cambridge


A woman sneezing into a tissue.


Given the year that was 2020, you’d be forgiven for forgetting that the flu was a big deal, we’ve had a lot on our plate virus-wise. But rates of flu have fallen dramatically, perhaps because of lockdown and social distancing measures. And while that may seem wonderful in the short term, it can have some negative effects down the road. Derek Smith from the University of Cambridge spoke to Adam Murphy about the potential severity of the next flu season...

Derek - Adam, we don't know how bad it is going to be or not. Like with many situations with what's going on with COVID-19, we've just never been in this situation before in our lifetimes, and in times when there has been such good recording of data and really knowing what is going on. We may have a relatively mild season if COVID continues and there is still social distancing going on, or if COVID circulates and somehow manages to competitively exclude flu. Or we may be in a situation where people haven't had much flu circulating for a couple of years, and it comes back with a vengeance and we have to be prepared for the latter. It's important that people continue to get the vaccines against the flu if they are at risk, but the truth is we don't know what's going to happen.

Adam - How do we usually vaccinate against the flu, because it's a bit of a moving target, isn't it?

Derek - Flu is a moving target from a scientific perspective. It's absolutely fascinating to study because of that. But of course from a public health perspective, this makes it an extremely difficult thing to vaccinate against, yeah. The flu evolves to escape immunity that we have, so it has this seemingly endless capacity to incrementally change and evade the immunity that we might have built up with prior infections against flu or prior vaccinations against flu. And so for people who are at risk for severe influenza, this is why the recommendations are that these individuals are vaccinated every year, because in most years we change at least one of the four strains of flu that's in the flu vaccine.

Adam - If it's going to change so much from year to year, how do we pin down any kind of correct answer for what a vaccine is going to look like?

Derek - Yeah, there is a really astonishing global network of people working in hospitals and GP offices who are noticing people who come in that look like they might have influenza taking throat swabs, sending those to a so-called World Health Organisation National Influenza Laboratory in their country. Those swabs are analysed there to see whether or not they really are flu, and if they are, then those samples are sent to one of five WHO so-called Collaborating Centres across the world. There's one in London, one in the United States, one in Tokyo, one in Melbourne, Australia, and one in Beijing, China. And those strains are analysed in a great deal of detail in terms of the sequence and in terms of whether or not their phenotype has changed to escape immunity. This is happening in almost 140 countries across the world. Thousands of people involved. And yeah, it's an amazing global operation. And much of that infrastructure that has been built up in those countries is the same sort of infrastructure that has been leveraged for tracking COVID

Adam - The rates have dropped though, so does that mean there's a lot fewer samples with which to try and make a prediction? Could that end up being a problem?

Derek - Yes, that's absolutely right. We rely on all of that global surveillance to see what is happening with the evolution of the viruses worldwide. And in a typical year, there are a small number of millions of samples that are taken in this way that I described throughout the world, and the viruses that filter through to the Collaborating Centres as being representative of what is circulating. It's typically in the region of about 10,000 per year, and this year it's in the hundreds, not 10,000 or so. So the vaccine choice this year we made with a much smaller number of viruses. There's still a lot of surveillance that is going on looking for flu, so we really do know that there is very little flu that is going on, and it has made the choice of which strains should go in the flu vaccine much more difficult. Nevertheless, it has been identified that there are some variants of one of the types of flu, the so-called H3 viruses that have emerged in a couple of countries and have spread internationally. And indeed, so the H3 component of the vaccine has been updated for this year's recommendation that took place in discussions over the last couple of weeks. And that is the decision announced last week.

Adam - And what does that uncertainty translate to in practical terms? Does it mean that the vaccine could miss the virus entirely? Or is it how effective the vaccine is? What could the knock on effects be?

Derek - These are so-called vaccine mismatches, and they do reduce the effectiveness of the vaccine. It's never the case with any vaccine, just like we know with the COVID vaccine, not everybody is going to be protected who is vaccinated, but a large proportion of them are. And when there is a vaccine mismatch like this, when the viruses evolved after we make the vaccine choice until the vaccine actually gets used, yes, then that vaccine mismatch reduces the effectiveness of the vaccine, which means that some people who wouldn't have normally noticed that they would have gotten flu will get flu. But there's three main categories of how the vaccine protects. The same with the COVID vaccine. Whether or not it protects against people feeling like they're ill - so protecting against infection - whether or not it protects against severe disease, and then whether or not it protects against death. And even in years where there is a vaccine mismatch, even though more people might not be protected from infection with flu, it's still in many cases going to be protecting them from severe disease and from death.


Add a comment